2 papers
cs.LG2026
HiFloat4 Format for Language Model Inference
Yuanyong Luo, Jing Huang, Yu Cheng +19
This paper introduces HiFloat4 (HiF4), a block floating-point data format tailored for deep learning. Each HiF4 unit packs 64 4-bit elements with 32 bits of shared scaling metadata…
cs.LG2024
Ascend HiFloat8 Format for Deep Learning
Yuanyong Luo, Zhongxing Zhang, Richard Wu +16
This preliminary white paper proposes a novel 8-bit floating-point data format HiFloat8 (abbreviated as HiF8) for deep learning. HiF8 features tapered precision. For normal value e…